CN116098650B

CN116098650B - Signal detection position detection method, device and computer equipment

Info

Publication number: CN116098650B
Application number: CN202310159515.5A
Authority: CN
Inventors: 刘怡升; 郑淳铭; 廖致霖
Original assignee: Interface Optoelectronics Shenzhen Co Ltd; Interface Technology Chengdu Co Ltd; General Interface Solution Ltd
Current assignee: Interface Optoelectronics Shenzhen Co Ltd; Interface Technology Chengdu Co Ltd; General Interface Solution Ltd
Priority date: 2023-02-23
Filing date: 2023-02-23
Publication date: 2024-12-13
Anticipated expiration: 2043-02-23
Also published as: CN116098650A

Abstract

The present application relates to a method, an apparatus, a computer device, a storage medium and a computer program product for detecting a signal detection position. The method comprises the steps of detecting a current detection position through an ultrasonic physiological signal meter probe to obtain a sound signal to be detected corresponding to the current detection position, carrying out digital signal processing on the sound signal to be detected to obtain two-dimensional audio characteristics of the sound signal to be detected, inputting the two-dimensional audio characteristics of the sound signal to be detected into a trained sound signal classification model, outputting probability values of the sound signal to be detected classified into preset sound signal types, determining the signal type of the sound signal to be detected according to the probability values, determining the current detection position to be the target detection position under the condition that the signal type is the target signal type, and pushing a correct prompt message for indicating the current detection position to be the target detection position. The method can assist the detector to find the correct detection position, thereby effectively improving the reliability of signal detection.

Description

Method and device for detecting signal detection position and computer equipment

Technical Field

The present application relates to the field of signal detection technology, and in particular, to a method, an apparatus, a computer device, a storage medium, and a computer program product for detecting a signal detection position.

Background

With the continuous development of ultrasonic technology, ultrasonic-based signal detection technology is becoming more and more mature. Currently, ultrasonic-based signal detection techniques have been widely used in medical diagnosis, such as detecting physiological and pathological information of a human body by an ultrasonic physiological signal meter, i.e., ultrasonic diagnosis.

The ultrasonic diagnosis method generally utilizes an ultrasonic physiological signal meter probe to transmit ultrasonic signals to a specific part of a human body, then displays the ultrasonic signals reflected back from the specific part of the human body through a display screen, and determines physiological conditions of invisible areas below the skin of the human body, such as the activity function of certain organs in the human body, whether the tissues and organs contain gas or liquid and the like according to the waveform, attenuation condition and other information of the reflected waves.

Because the ultrasonic-based physiological signal detection has the advantages of painless, no damage, safety, convenience, real-time imaging, clear imaging and the like, the ultrasonic-based physiological signal detection is an indispensable diagnosis method in clinical medicine at present. However, in the conventional ultrasonic physiological signal detection, a detector often determines the position of the probe of the ultrasonic physiological signal meter by experience or sense, and when the experience of the detector is insufficient, it is difficult to find an accurate detection position, and meanwhile, the detection result is easily interfered by external noise, so that the defect of unreliable signal detection exists.

Disclosure of Invention

In view of the foregoing, it is desirable to provide a signal detection position detection method, apparatus, computer device, computer-readable storage medium, and computer program product that can improve signal detection reliability.

In a first aspect, the present application provides a method for detecting a signal detection position. The detection method comprises the following steps:

detecting the current detection position through an ultrasonic physiological signal meter probe to obtain a sound signal to be detected corresponding to the current detection position;

Carrying out digital signal processing on the sound signal to be detected to obtain the two-dimensional audio characteristics of the sound signal to be detected;

Inputting the two-dimensional audio characteristics of the sound signals to be tested into a trained sound signal classification model, outputting probability values of the sound signals to be tested classified into preset sound signal types, and determining the signal types of the sound signals to be tested according to the probability values;

and under the condition that the signal type is the target signal type, determining the current detection position as the target detection position, and pushing a correct prompt message for indicating that the current detection position is the target detection position.

In one embodiment, the training process of the sound signal classification model comprises the following steps:

Acquiring a plurality of types of sample noise signals and sample ultrasonic signals reflected back based on a target detection position, and determining respective corresponding sample tags according to the signal types of the sample noise signals and the sample ultrasonic signals;

Acquiring two-dimensional audio features corresponding to the sample noise signals and the sample ultrasonic signals respectively, taking the two-dimensional audio features with sample labels as a sample data set, and dividing the sample data set into a training data set and a test data set;

Inputting the two-dimensional audio features in the training data set into a sound signal classification model, and updating parameters of the sound signal classification model according to the difference between the prediction label and the sample label output by the sound signal classification model;

And verifying and evaluating the sound signal classification model by adopting the test data set to obtain the trained sound signal classification model.

In one embodiment, the detecting the current detection position by the ultrasonic physiological signal meter probe to obtain the sound signal to be detected corresponding to the current detection position includes:

transmitting an ultrasonic signal to the current detection position through an ultrasonic physiological signal meter probe;

And receiving the sound signal reflected by the current detection position, and taking the reflected sound signal as the sound signal to be detected.

In one embodiment, performing digital signal processing on a sound signal to be detected to obtain a two-dimensional audio feature of the sound signal to be detected, including:

Sampling, quantizing and encoding the sound signal to be detected to obtain a corresponding digital signal;

and extracting the audio characteristics of the digital signals to obtain the two-dimensional audio characteristics of the sound signals to be detected.

In one embodiment, after determining the signal type of the sound signal to be measured according to the probability value, the method further includes:

Under the condition that the signal type is not the target signal type, determining that the current detection position is not the target detection position, and pushing an error prompt message for indicating that the current detection position is not the target detection position;

And repeatedly executing the detection position of the ultrasonic physiological signal meter probe until the sound signal type corresponding to the detection position after the movement is the target signal type, determining the detection position after the movement as the target detection position, and pushing a correct prompt message for indicating that the detection position after the movement is the target detection position.

In one embodiment, the sound signal classification model is a convolutional neural network model, and sequentially comprises an input layer, a plurality of groups of combination layers formed by a two-dimensional convolutional layer and a maximum pooling layer, a flattening layer, a plurality of groups of combination layers formed by a full-connection layer and a random inactivation layer, a full-connection layer and an output layer.

In a second aspect, the application further provides a device for detecting the signal detection position. The detection device includes:

The signal acquisition module is used for detecting the current detection position through the ultrasonic physiological signal meter probe to acquire a sound signal to be detected corresponding to the current detection position;

the signal processing module is used for carrying out digital signal processing on the sound signal to be detected and obtaining the two-dimensional audio characteristics of the sound signal to be detected;

The signal classification module is used for inputting the two-dimensional audio characteristics of the sound signal to be detected into the trained sound signal classification model, outputting probability values of the sound signal to be detected classified into preset sound signal types, and determining the signal types of the sound signal to be detected according to the probability values;

And the message pushing module is used for determining the current detection position as the target detection position under the condition that the signal type is the target signal type, and pushing a correct prompt message for indicating the current detection position as the target detection position.

In a third aspect, the present application also provides a computer device. The computer device comprises a memory storing a computer program and a processor which when executing the computer program performs the steps of:

In a fourth aspect, the present application also provides a computer-readable storage medium. The computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of:

In a fifth aspect, the present application also provides a computer program product. The computer program product comprises a computer program which, when executed by a processor, implements the steps of:

The method, the device, the computer equipment, the storage medium and the computer program product for detecting the signal detection position detect the current detection position through the ultrasonic physiological signal meter probe to obtain a sound signal to be detected corresponding to the current detection position, perform digital signal processing on the sound signal to be detected to obtain two-dimensional audio characteristics of the sound signal to be detected, input the two-dimensional audio characteristics of the sound signal to be detected into a trained sound signal classification model, output probability values of the sound signal to be detected classified into preset sound signal types, determine the signal type of the sound signal to be detected according to the probability values, determine the current detection position to be the target detection position under the condition that the signal type is the target signal type, and push a correct prompt message for indicating that the current detection position is the target detection position. In the whole detection process of the signal detection position, the two-dimensional audio characteristics of the sound signal to be detected are input into the sound signal classification model, the signal type of the sound signal to be detected is obtained, the current detection position is determined to be the target detection position according to the signal type, and the prompt message for the correct detection position is correspondingly pushed, so that a detector is assisted to find the correct detection position, and the reliability of signal detection can be effectively improved.

Drawings

FIG. 1 is a diagram of an application environment of a method for detecting a signal detection position in one embodiment;

FIG. 2 is a flow chart of a method for detecting a signal detection position in one embodiment;

FIG. 3 is a flow diagram of training a classification model of an acoustic signal in one embodiment;

FIG. 4 is a flow chart illustrating measurement in combination with classification results of a sound signal classification model according to an embodiment;

FIG. 5 is a block diagram of a sound signal classification model in one embodiment;

FIG. 6 is a flow chart of a method for detecting a signal detection position according to another embodiment;

FIG. 7 is a block diagram of a signal detection device in one embodiment;

fig. 8 is an internal structural diagram of a computer device in one embodiment.

Detailed Description

The present application will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present application more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.

The method for detecting the signal detection position provided by the embodiment of the application can be applied to an application environment shown in fig. 1. Wherein the terminal 102 communicates with the server 104 via a network. The data storage system may store data that the server 104 needs to process. The data storage system may be integrated on the server 104 or may be located on a cloud or other network server. The user operates on the terminal 102 side, and the terminal 102 performs detection of the signal detection position in response to the user operation.

Specifically, the terminal 102 detects a current detection position through an ultrasonic physiological signal meter probe to obtain a sound signal to be detected corresponding to the current detection position, performs digital signal processing on the sound signal to be detected to obtain a two-dimensional audio feature of the sound signal to be detected, inputs the two-dimensional audio feature of the sound signal to be detected to a trained sound signal classification model, outputs probability values of the sound signal to be detected classified into preset sound signal types, determines the signal type of the sound signal to be detected according to the probability values, determines the current detection position as the target detection position under the condition that the signal type is the target signal type, and pushes a correct prompt message for indicating that the current detection position is the target detection position. The terminal 102 may be, but not limited to, various personal computers, notebook computers, smartphones, tablet computers, internet of things devices, portable wearable devices, and the internet of things devices may be an internet of things intelligent oscilloscope. The portable wearable device may be a smart watch, smart bracelet, headset, or the like. The server 104 may be implemented as a stand-alone server or as a server cluster of multiple servers.

In one embodiment, as shown in fig. 2, a method for detecting a signal detection position is provided, and the method is applied to the terminal 102 in fig. 1 for illustration, and includes the following steps:

And S100, detecting the current detection position through an ultrasonic physiological signal meter probe to obtain a sound signal to be detected corresponding to the current detection position.

The current detection position refers to a currently detected human body part such as abdomen, face, brain and the like, and the sound signal to be detected refers to a sound signal reflected by the current detection position.

Alternatively, a specific ultrasonic signal is transmitted to a currently detected human body part (abdomen, face, brain, etc.) through an ultrasonic physiological signal meter probe, and a sound signal returned from the currently detected human body part is received, with the reflected sound signal being taken as a sound signal to be measured.

And S200, carrying out digital signal processing on the sound signal to be detected to obtain the two-dimensional audio characteristics of the sound signal to be detected.

Wherein the sound signal to be measured is an ultrasonic continuous signal in an analog form.

Optionally, an Analog-to-Digital Converter (ADC) converts the returned ultrasonic continuous signal in Analog form into an ultrasonic discrete signal in digital form, and finally, the ultrasonic discrete signal in digital form is transmitted to an intelligent device terminal such as a mobile phone, and the intelligent device terminal such as the mobile phone performs spectrum analysis and mel frequency cepstrum coefficient (Mel Frequency Cepstrum Coefficient, MFCC) and other acoustic signal processing on the ultrasonic discrete signal in digital form to obtain the two-dimensional audio characteristics of the acoustic signal to be measured.

S300, inputting the two-dimensional audio characteristics of the sound signal to be tested into the trained sound signal classification model, outputting probability values of the sound signal to be tested classified into preset sound signal types, and determining the signal types of the sound signal to be tested according to the probability values.

The number of the preset sound signal types can be set according to actual conditions, for example, the preset sound signal types can be respectively set to be friction noise type, background noise type and target signal type returned by the target detection position.

Optionally, the two-dimensional audio features of the sound signal to be measured are input into a pre-trained convolutional neural network model, and probability values of the sound signal to be measured classified as each preset sound signal type are output. The range of the output probability value is [0,1] and the sum of the output probability values is 1, and the preset sound signal type corresponding to the maximum probability value is used as the signal type of the sound signal to be detected. Taking the predicted sound signal type as the friction noise type, the background noise type and the target signal type as examples, assuming that the probability value of the sound signal to be measured classified as the friction noise type is 0.2, the probability value of the sound signal to be measured as the background noise type is 0.1 and the probability value of the sound signal to be measured as the target signal type is 0.7, the target signal type corresponding to the highest probability value is taken as the signal type of the sound signal to be measured.

And S400, under the condition that the signal type is the target signal type, determining the current detection position as the target detection position, and pushing a correct prompt message for indicating that the current detection position is the target detection position.

Optionally, under the condition that the type of the sound signal to be detected output by the sound signal classification model is the target signal type, determining the current detection position as the target detection position, and displaying a correct prompt message that the current detection position is the target detection position on intelligent terminal equipment such as a mobile phone. The intelligent terminal equipment can provide the prompt message with correct current detection position in an acousto-optic (voice prompt and indicator light) or display (picture display and text display) mode. For example, the terminal device may be provided with two kinds of red and green indicator lamps, when the green light is turned on, the current detection position is correct, and when the current detection position is the target detection position, the green light on the terminal device is turned on, and the "current detection position is correct | please start measurement" is displayed.

The method comprises the steps of detecting a current detection position through an ultrasonic physiological signal meter probe to obtain a sound signal to be detected corresponding to the current detection position, carrying out digital signal processing on the sound signal to be detected to obtain two-dimensional audio characteristics of the sound signal to be detected, inputting the two-dimensional audio characteristics of the sound signal to be detected into a trained sound signal classification model, outputting probability values of the sound signal to be detected classified into preset sound signal types, determining the signal type of the sound signal to be detected according to the probability values, determining the current detection position as the target detection position under the condition that the signal type is the target signal type, and pushing a correct prompt message for indicating that the current detection position is the target detection position. In the whole detection process of the signal detection position, the two-dimensional audio characteristics of the sound signal to be detected are input into the sound signal classification model, the signal type of the sound signal to be detected is obtained, the current detection position is determined to be the target detection position according to the signal type, and the prompt message for the correct detection position is correspondingly pushed, so that a detector is assisted to find the correct detection position, and the reliability of signal detection can be effectively improved.

In one embodiment, as shown in fig. 3, the training process of the acoustic signal classification model includes:

S210, acquiring a plurality of types of sample noise signals and sample ultrasonic signals reflected back based on target detection positions, and determining respective corresponding sample tags according to the signal types of the sample noise signals and the sample ultrasonic signals;

s220, acquiring two-dimensional audio features corresponding to the sample noise signals and the sample ultrasonic signals, taking the two-dimensional audio features with the sample labels as a sample data set, and dividing the sample data set into a training data set and a test data set;

S230, inputting two-dimensional audio features in the training data set into a sound signal classification model, and updating parameters of the sound signal classification model according to the difference between a prediction label and a sample label output by the sound signal classification model;

And S240, adopting the test data set to verify and evaluate the sound signal classification model, and obtaining the trained sound signal classification model.

Firstly, various noises such as background noise, friction noise and the like possibly generated in a signal detection process are obtained according to actual conditions, and sample ultrasonic signals reflected by target detection positions are determined according to signal types of the various sample noise signals and the sample ultrasonic signals. And then, carrying out digital signal processing on the sound signal marked with the sample tag, converting the acquired sound signal into two-dimensional audio features, taking the two-dimensional audio features with the sample tag as a sample data set, and randomly dividing the sample data set into a training data set and a test data set according to a preset proportion.

The method comprises the steps of inputting two-dimensional audio features in a training data set into a sound signal classification model, calculating a loss function value by utilizing the difference between a model output value and a sample label, updating parameters of the sound signal classification model by using the loss function value, thereby training the sound signal classification model, importing a test data set into the sound signal classification model with updated parameters, verifying and evaluating the sound signal classification model, and obtaining the trained sound signal classification model after the verification and evaluation are passed.

In this embodiment, multiple types of sample noise signals and sample ultrasonic signals reflected back based on target detection positions are acquired, sample tags and two-dimensional audio features corresponding to the acquired sound signals are determined, the two-dimensional audio features with the sample tags are used as sample data sets, and the sample data sets are randomly divided into training data sets and test data sets, so that a trained sound signal classification model is acquired, the type of the sound signal to be detected is determined subsequently, and the reliability of signal detection is improved.

In one embodiment, detecting the current detection position by the ultrasonic physiological signal meter probe to obtain a sound signal to be detected corresponding to the current detection position includes:

In general, a specific detection method of an ultrasonic physiological signal meter is to use a probe having an ultrasonic signal emitting device to emit an ultrasonic signal to a region below the skin of a human body, and then receive the reflected sound signal, so as to determine the shape and characteristics of a non-visible region below the skin of the human body, thereby performing various medical applications. However, the specific detection angle, position, moving distance, etc. of the ultrasonic physiological signal meter probe often needs to depend on the experience of the tester, and when the experience of the tester is insufficient, the received echo signal (reflected sound signal) is easily interfered by friction noise and other background noise generated by the mobile device, so that the detected signal waveform is difficult to distinguish.

Based on the signal detection position detection method provided by the embodiment, when the ultrasonic physiological signal meter is utilized to detect signals, the reflected sound signals can be used as sound signals to be detected instead of being further analyzed and displayed, the type of the sound signals to be detected is judged through the sound signal classification model, and when the type of the sound signals to be detected is the target signal type, a prompt message with correct current detection position is pushed, so that a detector is assisted to find the correct detection position.

In this embodiment, after receiving the sound signal reflected by the current detection position, the reflected sound signal is used as a sound signal to be detected, the signal type of the sound signal to be detected is determined based on the sound signal classification model, and under the condition that the current detection position is the target detection position, the corresponding current detection position prompt message is pushed, so that a person with insufficient experience can be assisted to find a correct detection position, and the efficiency and reliability of signal detection are improved.

For the received sound signal to be detected, firstly, the returned ultrasonic continuous signal in an analog form is subjected to processing such as sampling, quantization, encoding and the like through an ADC, the ultrasonic continuous signal is converted into an ultrasonic discrete signal in a digital form, then the ultrasonic discrete signal in the digital form is transmitted to an intelligent equipment terminal such as a mobile phone, and the intelligent equipment terminal carries out sound wave signal processing such as frequency spectrum analysis, MFCC feature extraction and the like on the ultrasonic discrete signal in the digital form to acquire the two-dimensional frequency spectrum feature of the sound signal to be detected.

In this embodiment, digital signal processing is performed on the received ultrasonic continuous signal in an analog form, firstly, an ultrasonic discrete signal in a digital form is obtained through ADC conversion and transmitted to an intelligent device terminal such as a mobile phone, and then, the intelligent device terminal further performs processing such as spectrum analysis and MFCC feature extraction on the ultrasonic discrete signal in the digital form to obtain a two-dimensional spectrum feature of the sound signal to be detected, thereby improving the accuracy of signal detection.

Under the condition that the type of the sound signal to be detected output by the sound signal classification model is not the target signal type, judging that the current detection position is not the target detection position, and displaying error prompt information that the current detection position is not the target detection position on intelligent terminal equipment such as a mobile phone. The intelligent terminal equipment can provide the prompt message with correct current detection position in an acousto-optic (voice prompt and indicator light) or display (picture display and text display) mode. For example, the intelligent terminal equipment can be provided with red and green indicator lamps, the red lamp indicates that the current detection position is wrong when the red lamp is lighted, and the green lamp indicates that the current detection position is correct when the green lamp is lighted. In case the current probe location is not the target probe location, a red light on the terminal device is lit and "current probe location error | please try to move the probe to other locations" is displayed. After the detection position of the ultrasonic physiological signal meter probe is repeatedly executed until the sound signal type corresponding to the detection position after the movement is the target signal type, the detection position after the movement is judged to be the target detection position, and meanwhile, the intelligent equipment terminal pushes a correct prompt message for indicating that the detection position after the movement is the target detection position, for example, a green light on the terminal equipment is turned on, and the current detection position is displayed to be correct | please start measurement.

Referring to fig. 4, in an actual application, after the to-be-detected sound signal corresponding to the current detection position is obtained, the classification result of the sound signal classification model and the waveform of the to-be-detected sound signal observed by the human eye may be combined to determine whether the current detection position is the target detection position. Under the condition that the current detection position is not the target detection position, the detector can continuously move the position of the ultrasonic physiological signal meter probe until the type of a sound signal returned by the moved detection position is the target signal type and the waveform of a sound signal to be detected observed by human eyes is the target signal waveform, and based on the moved correct position, the measurement is started.

In the embodiment, aiming at the condition that the signal type output by the sound signal classification model is not the target signal type, an error prompt message for indicating that the current detection position is not the target detection position is pushed to prompt a detector to try to move the probe to other positions, and when the sound signal type returned by the moved detection position is the target signal type, the intelligent equipment terminal pushes a correct prompt message for indicating that the current detection position is the target detection position to assist the detector to measure at the correct detection position, so that the reliability of signal detection is effectively improved.

In one embodiment, the sound signal classification model is a convolutional neural network model, and sequentially comprises an input layer, a plurality of groups of combined layers formed by a two-dimensional convolutional layer and a maximum pooling layer, a flattening layer, a plurality of groups of combined layers formed by a full-connection layer and a random inactivation layer, a full-connection layer and an output layer.

Referring to fig. 5, the sound signal classification model is composed of an input layer, n groups of combination layers composed of a two-dimensional convolution layer and a maximum pooling layer, a flattening layer, n groups of combination layers composed of a full connection layer and a random inactivation layer, a full connection layer and an input layer. The activation function used by the acoustic signal model may be ReLU, sigmoid, tanh, softmax or the like. For example, the output value of the fully connected layer may be processed using a softmax function to convert the output value to a probability value. The output probability value ranges are [0,1] and the sum of the output probability values is 1, and the sound signal classification model takes the preset sound signal type corresponding to the maximum probability value as the signal type of the sound signal to be detected.

In this embodiment, a convolutional neural network model composed of one input layer, n groups of combination layers composed of a two-dimensional convolutional layer and a maximum pooling layer, one flattening layer, n groups of combination layers composed of a full-connection layer and a random inactivation layer, one full-connection layer, and one input layer is set as the acoustic signal classification model. By constructing a convolutional neural network model, the input two-dimensional audio features are analyzed and processed, so that the signal type of the sound signal to be detected is obtained, a detector is assisted to find a correct detection position, and the reliability of signal detection is effectively improved.

In order to describe the technical solution of the signal detection position detection method of the present application in detail, a specific application example will be adopted in the following, and the whole processing procedure will be described with reference to fig. 6, which specifically includes the following steps:

1. and detecting the current detection position through the ultrasonic physiological signal meter probe to obtain a sound signal to be detected corresponding to the current detection position. The method specifically comprises the following steps:

a) The ultrasonic physiological signal meter probe transmits a specific ultrasonic signal to the currently detected human body parts (abdomen, face, brain, etc.).

B) And receiving the sound signal returned by the currently detected human body part, and taking the reflected sound signal as the sound signal to be detected.

2. And carrying out digital signal processing on the reflected sound signal to be detected to obtain the two-dimensional audio characteristics of the sound signal to be detected. The method specifically comprises the following steps:

a) The reflected analog form of the sound signal to be detected is sampled, quantized, encoded and the like through the ADC, and is converted into the digital form of ultrasonic discrete signal.

B) Transmitting the ultrasonic discrete signals in the digital form to intelligent equipment terminals such as mobile phones and the like, and performing sound wave signal processing such as frequency spectrum analysis, MFCC feature extraction and the like on the ultrasonic discrete signals in the digital form by the intelligent equipment terminals to obtain two-dimensional frequency spectrum features of the sound signals to be detected.

3. And training a sound signal classification model. The sound signal classification model is a convolutional neural network model, and sequentially comprises an input layer, a plurality of groups of combined layers formed by a two-dimensional convolutional layer and a maximum pooling layer, a flattening layer, a plurality of groups of combined layers formed by a full-connection layer and a random inactivation layer, a full-connection layer and an output layer. The specific process of training the sound signal classification model comprises the following steps:

a) And acquiring a friction noise signal, a background noise signal and a sample ultrasonic signal reflected by the target detection position, which are generated in the signal detection process, and determining respective corresponding sample labels according to signal types of the friction noise signal, the background noise signal and the sample ultrasonic signal.

B) And carrying out digital signal processing on the sound signal marked with the sample tag, converting the acquired sound signal into a two-dimensional audio feature, taking the two-dimensional audio feature with the sample tag as a sample data set, and randomly dividing the sample data set into a training data set and a test data set according to a preset proportion.

C) And inputting the two-dimensional audio features in the training data set into the sound signal classification model, calculating a loss function value by utilizing the difference between the model output value and the sample label, and updating the parameters of the sound signal classification model by applying the loss function value so as to train the sound signal classification model.

D) And importing the test data set into the voice signal classification model with updated parameters, performing verification and evaluation on the voice signal classification model, and obtaining the trained voice signal classification model after the verification and evaluation pass.

4. Inputting the two-dimensional audio characteristics of the sound signal to be tested into the trained sound signal classification model, outputting probability values of the sound signal to be tested classified into preset sound signal types (friction noise signal type, background noise signal type and target signal type), and determining the signal type of the sound signal to be tested according to the probability values.

5. Judging whether the current detection position is the target detection position according to the signal type of the sound signal to be detected, and correspondingly pushing the prompt message in an acousto-optic (voice prompt and indicator light) or display (picture display and text display) mode. The method specifically comprises the following steps:

a) And under the condition that the signal type is the target signal type, determining the current detection position as the target detection position, and pushing a correct prompt message for indicating that the current detection position is the target detection position. For example, a green light on the terminal device lights up and displays "current detected position correct | please start measurement".

B) And under the condition that the signal type is not the target signal type, determining that the current detection position is not the target detection position, and pushing an error prompt message for indicating that the current detection position is not the target detection position. For example, a red light on the terminal device lights up and displays "current detected position error | please try to move the probe to another position". In the case that the type of the sound signal returned by the moved detection position is the target signal type, determining that the corresponding moved detection position is the target detection position, pushing a correct prompt message for indicating that the corresponding moved detection position is the target detection position, for example, lighting a green light on the terminal device, and displaying a 'current detection position is correct | please start measurement'.

It should be understood that, although the steps in the flowcharts related to the embodiments described above are sequentially shown as indicated by arrows, these steps are not necessarily sequentially performed in the order indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in the flowcharts described in the above embodiments may include a plurality of steps or a plurality of stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of the steps or stages is not necessarily performed sequentially, but may be performed alternately or alternately with at least some of the other steps or stages.

Based on the same inventive concept, as shown in fig. 7, the embodiment of the present application further provides a signal detection position detection device for implementing the above-mentioned signal detection position detection method. The detection device comprises:

the signal acquisition module 701 is configured to detect a current detection position through the ultrasonic physiological signal meter probe, and acquire a sound signal to be detected corresponding to the current detection position;

The signal processing module 702 is configured to perform digital signal processing on the sound signal to be detected, and obtain a two-dimensional audio feature of the sound signal to be detected;

The signal classification module 703 is configured to input the two-dimensional audio feature of the to-be-detected sound signal to the trained sound signal classification model, output probability values of the to-be-detected sound signal classified into each preset sound signal type, and determine the signal type of the to-be-detected sound signal according to the probability values;

The message pushing module 704 is configured to determine that the current detection position is the target detection position if the signal type is the target signal type, and push a correct prompt message for indicating that the current detection position is the target detection position.

The signal detection position detection device detects the current detection position through the ultrasonic physiological signal meter probe to obtain a sound signal to be detected corresponding to the current detection position, performs digital signal processing on the sound signal to be detected to obtain two-dimensional audio characteristics of the sound signal to be detected, inputs the two-dimensional audio characteristics of the sound signal to be detected into a trained sound signal classification model, outputs probability values of the sound signal to be detected classified into preset sound signal types, determines the signal type of the sound signal to be detected according to the probability values, determines the current detection position as the target detection position under the condition that the signal type is the target signal type, and pushes a correct prompt message for indicating that the current detection position is the target detection position. In the whole detection process of the signal detection position, the two-dimensional audio characteristics of the sound signal to be detected are input into the sound signal classification model, the signal type of the sound signal to be detected is obtained, the current detection position is determined to be the target detection position according to the signal type, and the prompt message for the correct detection position is correspondingly pushed, so that a detector is assisted to find the correct detection position, and the reliability of signal detection can be effectively improved.

In one embodiment, the signal classification module 703 is further configured to obtain multiple types of sample noise signals and sample ultrasonic signals reflected back based on the target detection position, determine respective corresponding sample tags according to signal types of the sample noise signals and the sample ultrasonic signals, obtain two-dimensional audio features of the sample noise signals and the sample ultrasonic signals, respectively, use the two-dimensional audio features with the sample tags as a sample data set, divide the sample data set into a training data set and a test data set, input the two-dimensional audio features in the training data set into a sound signal classification model, update parameters of the sound signal classification model according to differences between the prediction tags and the sample tags output by the sound signal classification model, and perform verification and evaluation on the sound signal classification model by using the test data set to obtain a trained sound signal classification model.

In one embodiment, the signal acquisition module 701 is further configured to send an ultrasonic signal to the current detection position through the ultrasonic physiological signal meter probe, receive the sound signal reflected by the current detection position, and use the reflected sound signal as the sound signal to be detected.

In one embodiment, the signal processing module 702 is further configured to sample, quantize, and encode the sound signal to be detected to obtain a corresponding digital signal, and perform audio feature extraction on the digital signal to obtain two-dimensional audio features of the sound signal to be detected.

In one embodiment, the message pushing module 704 is further configured to determine that the current detection position is not the target detection position if the signal type is not the target signal type, and push an error prompt message for indicating that the current detection position is not the target detection position, and repeatedly execute moving the detection position of the ultrasonic physiological signal meter probe until the sound signal type corresponding to the moved detection position is the target signal type, determine that the corresponding moved detection position is the target detection position, and push a correct prompt message for indicating that the corresponding moved detection position is the target detection position.

In one embodiment, the signal classification module 703 is further configured to construct a sound signal classification model, where the sound signal classification model includes an input layer, a plurality of groups of combination layers including a two-dimensional convolution layer and a max-pooling layer, a flattening layer, a plurality of groups of combination layers including a full-connection layer and a random inactivation layer, a full-connection layer, and an output layer in order.

The above-mentioned respective modules in the signal detection position detection device may be implemented in whole or in part by software, hardware, and combinations thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.

In one embodiment, a computer device is provided, which may be a terminal, and the internal structure thereof may be as shown in fig. 8. The computer device includes a processor, a memory, an input/output interface, a communication interface, a display unit, and an input means. The processor, the memory and the input/output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input/output interface. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The input/output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless mode can be realized through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. The computer program is executed by a processor to implement a method of detecting a signal detection position. The display unit of the computer device is used for forming a visual picture, and can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, can also be a key, a track ball or a touch pad arranged on the shell of the computer equipment, and can also be an external keyboard, a touch pad or a mouse and the like.

It will be appreciated by those skilled in the art that the structure shown in FIG. 8 is merely a block diagram of some of the structures associated with the present inventive arrangements and is not limiting of the computer device to which the present inventive arrangements may be applied, and that a particular computer device may include more or fewer components than shown, or may combine some of the components, or have a different arrangement of components.

In one embodiment, a computer device is provided comprising a memory and a processor, the memory having stored therein a computer program, the processor when executing the computer program performing the steps of:

In one embodiment, the processor further performs the steps of acquiring a plurality of types of sample noise signals and sample ultrasonic signals reflected back based on the target detection positions, determining respective corresponding sample tags according to respective signal types of the sample noise signals and the sample ultrasonic signals, acquiring two-dimensional audio features of the sample noise signals and the sample ultrasonic signals, taking the two-dimensional audio features with the sample tags as a sample data set, dividing the sample data set into a training data set and a test data set, inputting the two-dimensional audio features in the training data set into a sound signal classification model, updating parameters of the sound signal classification model according to differences between predicted tags and the sample tags output by the sound signal classification model, and performing verification and evaluation on the sound signal classification model by using the test data set to acquire a trained sound signal classification model.

In one embodiment, the processor when executing the computer program further performs the steps of transmitting an ultrasonic signal to the current probe location through the ultrasonic physiological signal meter probe, receiving a sound signal reflected back from the current probe location, and taking the reflected sound signal as the sound signal to be measured.

In one embodiment, the processor performs the steps of sampling, quantizing and encoding the sound signal to be detected to obtain a corresponding digital signal, and performing the audio feature extraction on the digital signal to obtain the two-dimensional audio feature of the sound signal to be detected.

In one embodiment, the processor when executing the computer program further performs the steps of determining that the current detection position is not the target detection position and pushing an error prompt message for indicating that the current detection position is not the target detection position if the signal type is not the target signal type, repeatedly executing the detection position of the moving ultrasonic physiological signal meter probe until the sound signal type corresponding to the moved detection position is the target signal type, determining that the corresponding moved detection position is the target detection position, and pushing a correct prompt message for indicating that the corresponding moved detection position is the target detection position.

In one embodiment, the processor when executing the computer program further implements the steps of constructing a sound signal classification model, the sound signal classification model being a convolutional neural network model, the sound signal classification model comprising, in order, an input layer, a plurality of groups of combined layers comprising a two-dimensional convolutional layer and a max-pooling layer, a flattening layer, a plurality of groups of combined layers comprising a fully-connected layer and a random inactivation layer, a fully-connected layer, and an output layer.

In one embodiment, a computer readable storage medium is provided having a computer program stored thereon, which when executed by a processor, performs the steps of:

In one embodiment, the computer program when executed by the processor further comprises the steps of obtaining a plurality of types of sample noise signals and sample ultrasonic signals reflected back based on target detection positions, determining respective corresponding sample tags according to signal types of the sample noise signals and the sample ultrasonic signals, obtaining two-dimensional audio features of the sample noise signals and the sample ultrasonic signals, taking the two-dimensional audio features with the sample tags as sample data sets, dividing the sample data sets into training data sets and test data sets, inputting the two-dimensional audio features in the training data sets into a sound signal classification model, updating parameters of the sound signal classification model according to differences between prediction tags and the sample tags output by the sound signal classification model, and verifying and evaluating the sound signal classification model by the test data sets to obtain the trained sound signal classification model.

In one embodiment, the computer program when executed by the processor further performs the steps of transmitting an ultrasonic signal to the current probe location via the ultrasonic physiological signal meter probe, receiving a sound signal reflected back from the current probe location, and taking the reflected sound signal as the sound signal to be measured.

In one embodiment, the computer program when executed by the processor further performs the steps of sampling, quantizing and encoding the sound signal to be tested to obtain a corresponding digital signal, and performing audio feature extraction on the digital signal to obtain two-dimensional audio features of the sound signal to be tested.

In one embodiment, the computer program when executed by the processor further performs the steps of determining that the current probe position is not the target probe position and pushing an error prompt message for indicating that the current probe position is not the target probe position if the signal type is not the target signal type, repeatedly executing the probe positions of the moving ultrasonic physiological signal meter probe until the sound signal type corresponding to the moved probe position is the target signal type, determining that the corresponding moved probe position is the target probe position, and pushing a correct prompt message for indicating that the corresponding moved probe position is the target probe position.

In one embodiment, the computer program when executed by the processor further implements the steps of constructing a sound signal classification model, the sound signal classification model being a convolutional neural network model, the sound signal classification model comprising, in order, an input layer, a plurality of sets of combined layers comprising a two-dimensional convolutional layer and a max-pooling layer, a flattening layer, a plurality of sets of combined layers comprising a fully-connected layer and a random inactivation layer, a fully-connected layer, and an output layer.

In one embodiment, a computer program product is provided comprising a computer program which, when executed by a processor, performs the steps of:

It should be noted that, the user information (including but not limited to user equipment information, user personal information, etc.) and the data (including but not limited to data for analysis, stored data, presented data, etc.) related to the present application are information and data authorized by the user or sufficiently authorized by each party, and the collection, use and processing of the related data need to comply with the related laws and regulations and standards of the related country and region.

Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, database, or other medium used in embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high density embedded nonvolatile Memory, resistive random access Memory (ReRAM), magneto-resistive random access Memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric Memory (Ferroelectric Random Access Memory, FRAM), phase change Memory (PHASE CHANGE Memory, PCM), graphene Memory, and the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, and the like. By way of illustration, and not limitation, RAM can be in various forms such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), etc. The databases referred to in the embodiments provided herein may include at least one of a relational database and a non-relational database. The non-relational database may include, but is not limited to, a blockchain-based distributed database, and the like. The processor referred to in the embodiments provided in the present application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic unit, a data processing logic unit based on quantum computing, or the like, but is not limited thereto.

The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.

The foregoing examples illustrate only a few embodiments of the application and are described in detail herein without thereby limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of the application should be assessed as that of the appended claims.

Claims

1. A method for detecting a signal detection position, the method comprising:

performing digital signal processing on the sound signal to be detected to obtain the two-dimensional audio characteristics of the sound signal to be detected;

inputting the two-dimensional audio characteristics of the sound signal to be tested into a trained sound signal classification model, outputting probability values of the sound signal to be tested classified into preset sound signal types, and determining the signal types of the sound signal to be tested according to the probability values;

2. The method of claim 1, wherein the training process of the acoustic signal classification model comprises:

acquiring a plurality of types of sample noise signals and sample ultrasonic signals reflected back based on a target detection position, and determining respective corresponding sample tags according to signal types of the sample noise signals and the sample ultrasonic signals;

Acquiring two-dimensional audio features corresponding to the sample noise signal and the sample ultrasonic signal respectively, taking the two-dimensional audio features with a sample tag as a sample data set, and dividing the sample data set into a training data set and a test data set;

Inputting the sample ultrasonic signals and the sample noise signals in the training data set into a sound signal classification model, and updating parameters of the sound signal classification model according to the difference between the prediction label output by the sound signal classification model and the sample label;

and adopting the test data set to verify and evaluate the sound signal classification model to obtain the trained sound signal classification model.

3. The method according to claim 1, wherein the detecting the current detection position by the ultrasonic physiological signal meter probe to obtain the sound signal to be detected corresponding to the current detection position includes:

and receiving the sound signal reflected by the current detection position, and taking the reflected sound signal as a sound signal to be detected.

4. The method according to claim 1, wherein the performing digital signal processing on the sound signal to be detected to obtain a two-dimensional audio feature of the sound signal to be detected includes:

5. The method according to claim 1, wherein after determining the signal type of the sound signal to be measured according to the probability value, further comprising:

6. The method according to any one of claims 1 to 5, wherein the acoustic signal classification model is a convolutional neural network model, and the acoustic signal classification model sequentially includes an input layer, a plurality of sets of combination layers consisting of a two-dimensional convolutional layer and a max-pooling layer, a flattening layer, a plurality of sets of combination layers consisting of a full-connection layer and a random inactivation layer, a full-connection layer, and an output layer.

7. A device for detecting a signal detection position, the device comprising:

The signal classification module is used for inputting the two-dimensional audio characteristics of the sound signal to be detected into a trained sound signal classification model, outputting probability values of the sound signal to be detected classified into preset sound signal types, and determining the signal types of the sound signal to be detected according to the probability values;

8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the detection method according to any one of claims 1 to 6 when the computer program is executed.

9. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the detection method according to any one of claims 1 to 6.

10. A computer program product comprising a computer program, characterized in that the computer program, when executed by a processor, implements the steps of the detection method according to any one of claims 1 to 6.